Journal of Neuroimmunology 276 (2014) 71–79

Contents lists available at ScienceDirect

Journal of Neuroimmunology journal homepage: www.elsevier.com/locate/jneuroim

Gold drug auranofin could reduce neuroinflammation by inhibiting microglia cytotoxic secretions and primed respiratory burst Jocelyn M. Madeira a, Ekta Bajwa a, Maegan J. Stuart a, Sadayuki Hashioka b,c, Andis Klegeris a,⁎ a b c

Department of Biology, University of British Columbia Okanagan Campus, Kelowna, BC, Canada Kinsmen Laboratory of Neurological Research, Department of Psychiatry, University of British Columbia, Vancouver, BC, Canada Department of Psychiatry, Faculty of Medicine, Shimane University, Izumo, Shimane, Japan

a r t i c l e

i n f o

Article history: Received 30 September 2013 Received in revised form 6 August 2014 Accepted 11 August 2014 Keywords: Alzheimer's disease Parkinson's disease Neuroprotection Aurothiomalate Aurothiosulfate Respiratory burst priming

a b s t r a c t Neuroinflammation contributes to the pathogenesis of neurological disorders. Anti-inflammatory treatments could potentially be used to slow down the progression of these diseases. We studied the antineuroinflammatory activity of gold compounds which have been used to treat rheumatoid arthritis. Non-toxic concentrations of auranofin (0.1–1 μM) significantly reduced the cytotoxic secretions by primary human microglia and microglia-like THP-1 promonocytic cells. Auranofin inhibited primed NADPH-oxidase dependent respiratory burst and secretion of tumor necrosis factor (TNF)-α and nitric oxide by monocytic cells. It had a direct neuroprotective effect on SH-SY5Y neuronal cells. Auranofin could have a novel application in the treatment of neurodegenerative diseases. © 2014 Elsevier B.V. All rights reserved.

1. Introduction In recent years, accumulating data have indicated that inflammation in the central nervous system (CNS) contributes to several neurological impairments including Alzheimer's and Parkinson's diseases (Zhang et al., 2005; Frohman et al., 2006; Lee et al., 2010). Currently, there are no effective clinical treatments to prevent or stop this inflammation; therefore, research into novel therapeutic approaches is warranted. Inflammation in the CNS is driven by two main glial cell types: microglia and astrocytes (Li et al., 2011; Zilka et al., 2012). Both these cell types could be targeted by therapies aimed at reducing neuroinflammation (Block et al., 2007). In their activated pro-inflammatory state, microglia and astrocytes increase the secretion of toxins and inflammatory mediators. This could lead to reduced viability of the healthy neurons surrounding activated glia (Cameron and Landreth, 2010). Pharmacological means could potentially be used to reduce the neuronal loss caused by neuroinflammation through either suppressing the release of neurotoxins from glia or inducing their secretion of neurotrophic factors. Substances that act on neurons directly protecting them from glial neurotoxins could also be beneficial in AD and other neuropathologies that are partially driven by neuroinflammation. ⁎ Corresponding author at: Department of Biology, University of British Columbia Okanagan Campus, 3187 University Way, Kelowna, BC V1V 1V7, Canada. Tel.: +1 250 807 9557; fax: +1 250 807 8830. E-mail address: [email protected] (A. Klegeris).

http://dx.doi.org/10.1016/j.jneuroim.2014.08.615 0165-5728/© 2014 Elsevier B.V. All rights reserved.

Rheumatoid arthritis is a chronic autoimmune disorder characterized by inflammation in the joints (Kean, 1990). Gold compounds have been used in the treatment of rheumatoid arthritis. In North America, clinically available gold-containing therapeutics include intramuscular injections of aurothiomalate (ATM) and the oral gold compound 2,3,4,6-tetra-o-acetyl-L-thio-β-D-glucopyrano-sato-S-(triethylphosphine) gold manufactured as auranofin (AF) (Champion et al., 1990; Kean, 1990; Kean et al., 1997). The exact mechanisms of AF's anti-inflammatory activity have not been established; though a range of different effects of AF on peripheral immune cells have been documented (Yamashita et al., 1997; Stern et al., 2005; Kim et al., 2007, 2010; Nakaya et al., 2011). AF affects cytokine levels by increasing secretion of interleukin (IL)-8 and reducing IL-6 secretion from lipopolysaccharide (LPS) stimulated human monocytic cells (Stern et al., 2005; Kim et al., 2007). AF also induces the anti-inflammatory enzyme heme oxygenase (HOX)-1 in monocytic THP-1 cells (Kim et al., 2010). The various mechanisms of AF action have been summarized in a recent review (Madeira et al., 2012a). The peripheral anti-inflammatory activity of gold compounds has been characterized extensively, whereas the effects of these compounds on neuroimmune reactions are less studied, even though AF has been shown to reach the CNS after its oral administration (Walz et al., 1983; Madeira et al., 2012a, 2013). In this study we used three different monocytic cell lines to model microglia as well as cultured primary human cells. Microglia represent the mononuclear phagocyte system in the CNS and therefore play a central role in neuroimmune reactions (Ransohoff and Brown, 2012).

72

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79

We investigated two of the clinically available drugs: ATM and AF, along with aurothiosulfate (ATS), which is a structurally similar monovalent gold thiol compound shown to be ineffective as an anti-inflammatory drug (Bruze et al., 1995). Compounds were tested for their ability to reduce the secretion of pro-inflammatory cytokines and cytotoxins produced by activated human monocytic cells. In addition, we investigated the effects of gold compounds on phagocyte respiratory burst and investigated the ability of the gold compounds to protect neuronal cells from toxicity induced by supernatants from stimulated microglia-like cells. Of the three gold compounds studied, only AF exhibited potentially anti-neurotoxic and neuroprotective activity. 2. Materials and methods 2.1. Reagents The following substances were obtained from Sigma-Aldrich (Oakville, ON, Canada): ATM, LPS (from Escherichia coli 055-B5), diaphorase (EC 1.8.1.3, from Clostridium kluyveri, 5.8 U/mg solid), dimethyl sulfoxide (DMSO), Triton X-100, luminol sodium salt, N-formyl-metleu-phe (fMLP), p-iodonitrotetrazolium violet, NAD+, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT), sulfanilamide and N-1-naphthylethylenediamine. AF was supplied by Cedarlane Canada (Burlington, ON, Canada). ATS was from VWR International (Mississauga, ON, Canada). Human and murine recombinant interferon (IFN)-γ, monocyte chemoattractant protein (MCP)-1 and tumor necrosis factor (TNF)-α enzyme linked immunosorbent assay (ELISA) kits were purchased from Peprotech (Rocky Hill, NJ, USA). All other reagents were obtained from Thermo Fisher Scientific (Nepean, ON, Canada) unless stated otherwise. 2.2. Cell culture The human monocytic THP-1 and promyelocytic HL-60 cell lines were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The human neuroblastoma SH-SY5Y cell line was a gift from Dr. R. Ross, Fordham University, NY. Murine BV-2 microglial cell line was a gift from Dr. G. Garden, University of Washington, WA. Human primary microglia were obtained from epileptic patients undergoing temporal lobe surgery. The specimens were from normal tissue overlying the epileptic foci. The use of human brain materials was approved by the Clinical Screening Committee for Human Subjects of the University of British Columbia. Microglia were isolated following protocols described by Hashioka et al. (2009). All cells were grown in Dulbecco's modified Eagle's medium-nutrient mixture F12 ham (DMEM-F12) supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 U/ml penicillin, 100 μg/ml streptomycin). The THP-1 and SH-SY5Y cell lines were used without differentiation; HL-60 cells were differentiated prior to use in experiments as described below. 2.3. Effects of compounds on cell viability and cytokine secretion Human monocytic THP-1 cells were seeded into 24-well plates at a concentration of 5 × 105 cells/ml in 0.9 ml of DMEM-F12 containing 5% FBS. Human microglia were plated at 7.5 × 104 cells per well in DMEM-F12 containing 5% FBS. Following 48 h incubation, media were removed and replaced with fresh medium. Cells were incubated in the presence of gold compounds or their vehicle solution (DMSO) for 15 min prior to the addition of the activating stimulus (0.5 μg/ml LPS plus 150 U/ml human recombinant IFN-γ). The final concentration of DMSO in cell culture medium did not exceed 0.13%; at this concentration, DMSO did not have any detectable effects on cell viability or function. After 24–48 h incubation, 0.1 ml of cell culture media was sampled for lactate dehydrogenase (LDH) enzymatic activity to determine the percentage of dead cells, while the evaluation of the surviving cells was performed by the MTT assay. The concentration of MCP-1 and

TNF-α (ng/ml) in THP-1 cell supernatants was measured in 0.1 ml of cell-free supernatants by ELISA according to the protocol provided by the supplier of antibodies (Peprotech). 2.4. Cytotoxicity of THP-1 cell and human microglia supernatants towards SH-SY5Y neuronal cells The experiments were performed as previously described (Klegeris et al., 2003). Briefly, THP-1 cells and primary human microglia were seeded into 24-well plates and stimulated in the presence or absence of various compounds as described above. After 24 h or 48 h incubation, 0.4 ml of cell-free supernatant was transferred to each well containing SH-SY5Y cells that had been plated 24 h earlier at a concentration of 2 × 105 cells/ml in 0.4 ml of DMEM-F12 medium containing 5% FBS. After 72 h incubation, neuronal cell death was assessed by the LDH assay and neuronal cell viability was assessed by the MTT assay. SHSY5Y cells in culture were also observed with an inverted phase contrast microscope (Motic, Richmond, BC, Canada) and photographed using a Motic 3000 digital camera. 2.5. Protective activity of AF on SH-SY5Y neuronal cells exposed to supernatants from stimulated THP-1 cells THP-1 cells were plated in 10 cm tissue culture plates at 2 × 105 cells/ml in 15 ml of DMEM-F12 containing 5% FBS. Following 15 min incubation, THP-1 cells were stimulated with a combination of 0.5 μg/ml LPS and 150 U/ml IFN-γ. After 24 h, 0.4 ml of cell-free supernatant was transferred to each well containing SH-SY5Y cells. Immediately after transfer of supernatants, SH-SY5Y cells were treated with AF or its vehicle solution (DMSO). Following 72 h incubation, the survival of neuronal cells was measured by the MTT assay. Control experiments were performed to evaluate the direct effects of AF on neuronal cell viability by treating SH-SY5Y cells grown in DMEM-F12 containing 5% FBS with AF or its vehicle solution (DMSO) for 72 h and evaluating neuronal cell viability by the MTT assay. 2.6. Cell viability assays: LDH release Cell death was evaluated by measuring LDH enzymatic activity in cell culture supernatants as described by Decker and LohmannMatthes (1988). In this assay, formation of the formazan product of iodonitrotetrazolium dye was followed colorimetrically. Briefly, 0.1 ml of cell culture supernatants was pipetted into the wells of 96-well plates, followed by an addition of 15 μl lactate (36 mg/ml) and 15 μl piodonitrotetrazolium violet (2 mg/ml) solutions. The enzymatic reaction was started by the addition of NAD +/diaphorase solution (3 mg/ml NAD+, 2.3 mg solid/ml diaphorase). After a 15–30 min incubation period, optical densities at 490 nm were measured by a FLUOstar Omega microplate reader (BMG Labtech, Offenburg, Germany). The amount of LDH which had been released was expressed as a fraction of the value obtained in comparative wells where the remaining cells were completely lysed by 1% Triton X-100. 2.7. Cell viability assay: reduction of formazan dye (MTT) The MTT assay was performed as previously described (Mosmann, 1983; Hansen et al., 1989). This assay is based on the ability of viable, but not dead, cells to convert the tetrazolium salt MTT to colored formazan. The viability of cultured cells was determined by adding MTT to reach a final concentration of 0.5 mg/ml. Following a 1–2 h incubation at 37 °C, the dark crystals which had formed were dissolved by adding an equal volume of SDS/DMF (20% sodium dodecyl sulfate, 50% N,N-dimethyl formamide, pH 4.7) to each well. The plates were then incubated at 37 °C for 3 h. Optical densities were measured at 570 nm using a microplate reader after transferring 0.1 ml aliquots to 96-well

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79

plates. The viable cell value was calculated as a percent of the value obtained from cells incubated with fresh medium only.

73

block design Analysis of Variance (ANOVA) was used, followed by Fisher's least significant difference (LSD) post-hoc test. A P value less than 0.05 was considered statistically significant.

2.8. Nitric oxide (NO) production by murine BV-2 microglial cells NO production was determined indirectly by measuring nitrite (NO− 2 ) accumulation using the Griess reagent (Ding et al., 1988). Murine microglial BV-2 cells were plated at 2 × 105 cells/ml in 0.5 ml DMEM-F12 medium containing 5% FBS and incubated for 24 h to allow for adherence of cells. The media was refreshed and cells were pre-treated with vehicle (0.1% DMSO) or varying AF concentrations (0.05–0.5 μM) for 15 min before stimulation with 0.5 μg/ml LPS and 150 U/ml murine IFN-γ. Cell-free culture media were collected 24 h later and nitrite measurements were performed by mixing equal volumes (50 μl) of culture medium and Griess reagent (1% sulfanilamide, 0.1% N-1-naphthylethylenediamine, 2.5% phosphoric acid) followed by absorbance measurement at 570 nm. Standard sodium nitrite solutions were used to calibrate absorbance readings.

3. Results 3.1. Effects of gold compounds on human THP-1 promonocytic cell viability and cytotoxic secretions Gold compounds (0.1–2.5 μM) were tested for their ability to inhibit human THP-1 promonocytic cell toxicity towards human neuronal SHSY5Y cells and results were compared to those obtained from samples treated with DMSO vehicle solution only. At the concentration used (b0.13%, v/v), DMSO alone had no detectable effects in the assays used (data not shown). THP-1 cells were treated with the compounds for 15 min prior to stimulation with LPS plus IFN-γ. Following a 24 h incubation period, viability of THP-1 cells was assessed using the MTT cell viability (Fig. 1B) and LDH cell death (Fig. 1C) assays, which showed no

2.9. Measuring unprimed respiratory burst by luminol-dependent chemiluminescence DMSO-differentiated human promyelocytic HL-60 cells were used to study the effects of AF on the phagocyte respiratory burst. These cells have been shown to express all subunits of phagocyte NADPH oxidase (Muranaka et al., 2005). Real-time chemiluminescence measurements were performed as described by Muranaka et al. (2005). First, HL-60 cells were differentiated by plating them into 10 cm tissue culture plates at a density of 2 × 105 cells/ml in DMEM-F12 medium containing 10% FBS and 1.3% DMSO. After a 5–7 day incubation period, which was confirmed in preliminary studies to enhance the respiratory burst activity significantly, HL-60 cells were washed and transferred into DMEMF12 medium without phenol red containing 5% FBS. Cells were seeded into a 96-well plate at a concentration of 1 × 106 cells/ml in 80 μl of medium and the plate was inserted into the FLUOstar Omega plate-reader. The respiratory burst response was measured after injecting luminol solution first, followed by a 7.5 min baseline recording and subsequent fMLP injection directly into the wells. The final reaction volume in the well was 0.1 ml, the final luminol concentration was 5 mM and fMLP was at 1 μM. Chemiluminescence response of HL-60 cells was recorded for 30 min and expressed as the area under the curve in Relative Light Units. Plates were maintained at 37 °C during all experiments. The effect of AF on respiratory burst of unprimed cells was studied by injecting luminol, followed by fMLP, 15 min after addition of AF to HL-60 cells. Concentration of AF vehicle solvent (DMSO) in control wells did not exceed 0.5%; at this concentration DMSO did not affect the chemiluminescent response of cells. 2.10. Measuring LPS-primed respiratory burst by luminol-dependent chemiluminescence In some experiments, the respiratory burst response of DMSOdifferentiated HL-60 cells was enhanced (primed) by adding 0.5 μg/ml LPS to HL-60 cells 24 h before the injection of luminol and fMLP. HL-60 cells were plated and chemiluminescence measurements were performed exactly as described above. The effect of AF on respiratory burst priming was studied after its addition to cells 15 min before the priming agent LPS (24 h treatment), or 30 min before injecting fMLP (30 min treatment). Concentration of AF vehicle solvent (DMSO) in control wells did not exceed 0.5%; at this concentration DMSO did not affect the chemiluminescent response of cells. 2.11. Statistical analyses Due to considerable variability in the absolute values obtained from independent experiments performed on different days, randomized

Fig. 1. AF reduces cytotoxicity of human THP-1 monocytic cells towards human neuroblastoma SH-SY5Y cells. THP-1 cells were pre-treated with various concentrations of the gold compounds or their vehicle solution (DMSO) for 15 min before stimulation with a combination of LPS (0.5 μg/ml) and IFN-γ (150 U/ml). After 24 h incubation, THP-1 cell viability was assessed by the MTT (B) and LDH (C) assays. The viability of SH-SY5Y neuronal cells exposed for 72 h to cell-free supernatants from stimulated THP-1 cells was assessed by the MTT assay (A). The horizontal dashed line (A) indicates viability of SH-SY5Y cells exposed to supernatants from unstimulated THP-1 cells. Data from 4 independent experiments are presented. The concentration-dependent effects of the compounds were assessed by the randomized block design ANOVA, followed by Fisher's LSD post-hoc test. *P b 0.05 significantly different from samples treated with the vehicle solution only.

74

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79

toxicity of AF, ATM or ATS towards stimulated THP-1 cells at the concentrations tested. After the 24 h stimulation period, the cell-free supernatants from THP-1 cell cultures were transferred to SH-SY5Y neuroblastoma cells to assess their cytotoxic effects. Following a 72 h incubation period with THP-1 supernatants, the viability of neuronal cells was assessed using the MTT assay (Fig. 1A). Supernatants from unstimulated THP-1 cells did not significantly affect the viability of SH-SY5Y cells (see dashed line in Fig. 1A). Transfer of cell-free supernatants from stimulated THP-1 cells to SH-SY5Y cells resulted in significantly reduced neuronal cell viability (Fig. 1A). A combination of LPS and IFN-γ was used to achieve maximal stimulation of cells (Klegeris et al., 2005). Fig. 1A shows that ATM and ATS did not exhibit anti-neurotoxic activity while AF (0.5–2 μM) inhibited secretion of toxins by stimulated THP-1 cells. This anti-neurotoxic activity of AF was not due to its toxicity towards THP-1 cells, but most likely due to specific inhibition of the cytotoxic secretions of THP-1 cells. 3.2. Effects of AF on primary human microglia cell viability and cytotoxic secretions The anti-neurotoxic activity of AF was also confirmed by using primary microglia cultures prepared from human surgical tissue samples. AF was tested at 0.1 μM and cell viability values were compared to those obtained from samples treated with DMSO vehicle solution only. Following 48 h incubation, the viability of primary human microglia was assessed using the MTT assay (Fig. 2B); at 0.1 μM, AF was not toxic to microglia. Cell-free supernatants from microglia cultures were transferred to SH-SY5Y neuroblastoma cells in order to assess possible cytotoxic effects. Following a 72 h incubation period with these supernatants, the viability of neuronal cells was measured using the MTT assay (Fig. 2A), which demonstrated that at 0.1 μM AF inhibited the

toxicity of stimulated human microglia towards neuronal cells (Fig. 2A), confirming results observed with the THP-1 cells which were used as microglia model. The morphology of SH-SY5Y cells incubated with supernatants of human microglia was also analyzed (Fig. 2C–E). The supernatants of activated microglia caused significant changes of cellular morphology with many cells showing bright and circularly shrunk cytoplasm (Fig. 2D). This was different from control samples exposed to vehicle solution only, which showed mostly pyramidal cell bodies (Fig. 2C). The adverse changes in SH-SY5Y cell morphology induced by stimulated microglia supernatants were considerably inhibited by pre-treatment with 0.1 μM AF (Fig. 2E). These observations were consistent with the results obtained by the MTT assay (Fig. 2A). 3.3. Direct neuroprotective effects of AF The direct neuroprotective effect of AF on cultured SH-SY5Y cells was assessed by adding AF (0.1–1 μM) to supernatants from stimulated THP-1 cells at the time of their transfer to SH-SY5Y cells (Fig. 3A). AF at 0.5 and 1 μM significantly increased viability of SH-SY5Y cells. Fig. 3B demonstrates that at the concentrations studied, AF had no direct effect on SH-SY5Y cell viability in the absence of THP-derived toxins. 3.4. Effects of gold compounds on the respiratory burst The three gold compounds; AF, ATM and ATS, were tested for their ability to inhibit the NADPH oxidase-dependent respiratory burst (data not shown for ATM and ATS). DMSO-differentiated human HL60 cells were used as a model of the microglia/phagocyte respiratory burst as THP-1 cells do not express the NADPH enzyme at levels high enough to generate detectable levels of reactive oxygen species (ROS) in our experiments (data not shown). Gold compounds were tested at

Fig. 2. AF reduces cytotoxicity of human microglia towards human neuroblastoma SH-SY5Y cells. Primary human microglia were pre-treated with 0.1 μM AF or its vehicle solution (DMSO) for 15 min before stimulation with a combination of LPS (0.5 μg/ml) and IFN-γ (150 U/ml). After 24 h incubation, microglial cell viability was assessed by the MTT assay (B). The viability of SH-SY5Y neuronal cells exposed for 72 h to supernatants from stimulated microglia was assessed by the MTT assay (A). Data from 3 independent experiments with cells obtained from two different surgical cases are presented. The effects of AF were assessed by the randomized block design ANOVA, followed by Fisher's LSD post-hoc test. *P b 0.05, significantly different from stimulated samples treated with the vehicle solution only. Phase contrast microscopy of SH-SY5Y cell cultures after incubation with supernatants from human microglia that were unstimulated (C) or stimulated with LPS plus IFN-γ in the absence (D) or presence of 0.1 μM AF (E). Photos are representative of 3 independent experiments. Magnification bar (=50 μm) in C is representative for all three panels.

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79

75

Fig. 3. AF protects SH-SY5Y neuronal cells against toxicity induced by supernatants from stimulated THP-1 cells. AF (0.1–1 μM) or its vehicle solution (DMSO) was added to supernatants from stimulated THP-1 cells at the time of their transfer to SH-SY5Y cell cultures (A). AF (0.1–1 μM), or its vehicle solution (DMSO), was added to unstimulated SH-SY5Y cells (B). Viability of neuronal cells was assessed by the MTT assay 72 h later. Data from 5 independent experiments are presented; the concentration-dependent effects of AF were assessed by the randomized block design ANOVA, followed by Fisher's LSD post-hoc test. **P b 0.01 significantly different from samples treated with the vehicle solution only.

concentrations ranging from 0.1–1 μM and results were compared to those obtained from samples treated with DMSO vehicle control only. At the concentration used (b0.13%, v/v), DMSO alone had no detectable effects on the chemiluminescent response of cells (data not shown). Fig. 4A shows that the addition of AF 24 h prior to HL-60 stimulation with fMLP caused significant reduction in the respiratory burst response and the effect was statistically significant at 1 μM. However, since 1 μM AF reduced HL-60 viability after 24 h incubation (Fig. 4C), this effect could be at least partially due to toxicity of AF towards HL-60 cells. A shorter, 30 min incubation period of HL-60 cells with AF prior to their stimulation with fMLP did not result in reduced chemiluminescent response (Fig. 4B), which indicated that AF at the concentrations studied did not interfere with the chemiluminescence measurements by, for example, reacting with luminol or reactive oxygen species. 30 min incubation with AF also did not lead to reduction in HL-60 cell viability (data not shown). The other gold compounds studied did not significantly affect the luminol-dependent chemiluminescence of HL-60 cells when added 24 h prior to stimulation (data not shown). 3.5. AF inhibits LPS-primed respiratory burst The ability of AF to inhibit priming of the respiratory burst of HL-60 cells induced by 24 h treatment with LPS was investigated using DMSOdifferentiated human HL-60 cells. AF was tested at concentrations ranging from 0.1–1 μM and results were compared to those obtained from control samples treated with the DMSO vehicle solution. Fig. 5A shows data obtained in an experiment where AF was added to HL-60 cells 15 min before their exposure to the priming agent (LPS). Following 24 h incubation, ROS production was induced by adding fMLP and an increase in luminol-dependent chemiluminescence signal was recorded. Incubation of DMSO-differentiated HL-60 cells with LPS for 24 h was found to increase their chemiluminescent response by as much as 300% (see dashed lines for values obtained from unprimed cells in the absence of AF, Fig. 5A, B). Addition of AF (0.1–1 μM) 15 min before the priming agent (LPS) significantly reduced the production of ROS (Fig. 5A). AF at a concentration as low as 0.1 μM showed significant

effect. Acute treatment of LPS-primed HL-60 cells with AF 30 min prior to stimulation with fMLP had no effect on ROS production (Fig. 5B). Viability measurements of HL-60 cells co-exposed to LPS and AF indicated that 24 h incubation with the highest concentration of AF (1 μM) was toxic to HL-60 cells (Fig. 5C); therefore, inhibition of the respiratory burst priming by 1 μM AF could be partially due to its toxicity towards HL-60, while such non-specific inhibition could not account for the reduction of the respiratory burst priming at lower AF concentrations (0.1 and 0.5 μM). 3.6. Effects of AF on the secretion of TNF-α, MCP-1 and NO by monocytic cells The effects of AF on the secretion of the pro-inflammatory cytokines TNF-α and MCP-1 by stimulated THP-1 cells were investigated by ELISA. Fig. 6A shows that stimulation of THP-1 cells with LPS plus IFN-γ increased the secretion of TNF-α. AF at 1 μM partially inhibited secretion of this cytokine (Fig. 6A). Similarly, LPS plus IFN-γ stimulation induced secretion of MCP-1 by THP-1 cells, but AF did not inhibit MCP-1 release at non-toxic concentrations (0.5–2 μM) (data not shown). Murine microglial BV-2 cells were used to study the effects of AF on NO production since these cells respond readily to LPS plus IFN-γ stimulation by releasing high concentrations of NO (Gibbson and Dragunow, 2006). Fig. 6B illustrates that stimulation of BV-2 cells with a combination of LPS and murine IFN-γ caused a significant increase in nitrite concentration. Pre-treatment of cells with AF (0.05–0.5 μM) suppressed this secretion in a concentration-dependent manner. AF at the concentrations studied did not affect viability of monocytic cells, and it had no effect on all three parameters studied in unstimulated cells (data not shown). 4. Discussion The gold compounds AF, ATM, and ATS were investigated for their potential to reduce harmful effects of neuroinflammation using in vitro models relevant to the inflammatory processes that occur

76

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79

Fig. 4. Only high concentrations of AF inhibit the chemiluminescent response of HL-60 cells to fMLP stimulation. DMSO-differentiated HL-60 cells were seeded into 96-well plates and pre-treated with various concentrations of AF or its vehicle solution (DMSO) for 24 h (A) or 30 min (B). Luminol-dependent chemiluminescence response of HL-60 cells was recorded for 30 min after injection of 1 μM fMLP. Viability of HL-60 cells at the end of the experiment (conditions shown in A) was assessed by the MTT assay (C). Data from 4 independent experiments are presented; the concentration-dependent effects of the compounds were assessed by the randomized block design ANOVA, followed by Fisher's LSD post-hoc test. *P b 0.05 significantly different from samples treated with the vehicle solution only.

during neurodegeneration. We studied the effects of gold-containing compounds at low micromolar concentrations, which in the case of AF have been shown to be attainable after its oral administration (Walz et al., 1983; Madeira et al., 2013). It was found that at non-toxic concentrations (0.1–2 μM) AF reduced cytotoxicity of both stimulated human THP-1 promonocytic cells (0.5–2 μM) and primary human microglia (0.1 μM) towards neuronal SH-SY5Y cells, indicating a potential antineurotoxic effect of AF. ATM and ATS, the other two gold compounds studied, were inactive. Previous research using similar assays has already shown that primary human cells could be more sensitive to anti-inflammatory drug treatment than cell lines; therefore, the lower effective concentration of AF in primary cells was not unexpected (Yamada et al., 1999; Polanski et al., 2011). Similarly, 0.1 μM AF inhibited toxicity of primary human astrocytes, while 1–5 μM concentrations of AF were needed to observe such inhibitory activity by using an astrocytoma cell line (Madeira et al., 2013).

Fig. 5. AF inhibits LPS-primed respiratory burst of DMSO-differentiated HL-60 cells. Cells were seeded into 96-well plates and their respiratory burst response primed by 24 h incubation with 0.5 μg/ml LPS, followed by injection of 1 μM fMLP which is a trigger of the respiratory burst. The concentration-dependent effects of AF were studied by adding this drug 15 min before HL-60 exposure to LPS (A) or 30 min before fMLP injection (B). Luminol-dependent chemiluminescence response of HL-60 cells in the presence or absence of AF was recorded for 30 min after injection of fMLP (A, B). The horizontal dashed lines indicate chemiluminescence values obtained from unprimed cells in the absence of AF (A, B). Viability of HL-60 cells at the end of the experiment (conditions shown in A) was assessed by the MTT assay (C). Data from 4-5 independent experiments are presented; the concentration-dependent effects of AF were assessed by the randomized block design ANOVA, followed by Fisher's LSD post-hoc test. *P b 0.05, **P b 0.01 significantly different from samples treated with the vehicle solution only.

To determine whether the observed anti-neurotoxic activity of AF was due to its interaction with monocytic THP-1 and microglial cells, or whether this effect was due to the transfer of some of the compound with the monocytic supernatants and its subsequent direct action on neuronal cells, experiments were performed where AF was applied directly to neuronal cells at the time of their exposure to cytotoxic supernatants from stimulated THP-1 cells. Treating SH-SY5Y neuronal cells with 0.5–1 μM AF protected them from toxicity induced by supernatants from stimulated THP-1 cells. A direct protective effect of AF (0.5–1 μM) on neuronal cells that had been exposed to toxic concentrations of hydrogen peroxide or astrocytic toxins was reported previously (Madeira et al., 2013). These observations indicate that AF may confer neuroprotection in addition to its better-known and accepted antiinflammatory properties. The direct protective activity of AF on neuronal cells was unlikely the sole mechanism responsible for the antineurotoxic activity of AF observed in our study since AF effectively inhibited primary human microglial toxic secretions at 0.1 μM, a concentration that was ineffective at rescuing neuronal cells from several different types of toxins (Fig. 3; Madeira et al., 2013). It is also important to note that the inhibitory effect of AF on monocytic cells was not due to its non-specific toxicity since both of the cell viability assays employed

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79

Fig. 6. AF inhibits TNF-α and NO secretion by monocytic cells. Human monocytic THP-1 cells (A) and mouse microglial BV-2 (B) cells were pre-treated with various concentrations of AF or the vehicle solution (DMSO) for 15 min before stimulation with a combination of LPS and either human (A) or murine (B) IFN-γ. Following a 48 h (A) or 24 h (B) incubation period, TNF-α concentration in culture supernatants was measured by ELISA (A), and nitrite concentration was measured by Griess reagent (B). Data from 4 (A) and 7 (B) independent experiments are presented. The horizontal dashed lines correspond to values obtained in supernatants from unstimulated cells. The concentrationdependent effect of AF was assessed by the randomized block design ANOVA, followed by Fisher's LSD post-hoc test. *P b 0.05, **P b 0.01 significantly different from samples treated with the vehicle solution only.

to monitor monocytic cell viability showed no toxic effects of AF at low micromolar concentrations. In addition, AF at the concentrations studied did not inhibit secretion of MCP-1 by THP-1 monocytic cells. MCP-1 is a chemotactic cytokine, which is sometimes used as an indicator of inflammatory activation of mononuclear phagocytes (Abramson and Gallin, 1990). A wide range of substances have been shown to contribute to the neurotoxic and cytotoxic activity of microglia; most likely a mixture of substances is responsible for such effects, the composition of which depends on the cell type and stimuli used (for review see Block et al., 2007; Madeira et al., 2012b). Our data indicate that AF at non-toxic low micromolar concentrations inhibits the secretion of two known microglial toxins, TNF-α and NO (Gibbson and Dragunow, 2006; Brown, 2010; Lull and Block, 2010), which could be the basis for the anti-neurotoxic activity of AF observed in this study. One of the proposed strategies for the treatment of neurodegenerative diseases is to enhance neuronal viability by increasing tolerance of neurons to oxidative damage and inflammatory toxins released by activated glial cells (Pocernich and Butterfield, 2012; Cornelius et al., 2013). Glial cells themselves release several endogenous neuroprotective molecules, including insulin and pro-insulin, which have received attention as potential therapeutic agents for preventing the neuronal death in Alzheimer's and Parkinson's diseases (Ellingsen et al., 2001; Block et al., 2007). In addition to endogenous molecules, synthetic drugs and purified compounds have been developed in an attempt to increase neuronal viability. For example, flavonoids, such as quercetin, have been recognized as neuroprotective molecules; Dajas et al. (2003) found that 25 μM quercetin protected neuronal cells against hydrogen peroxide toxicity. This protective effect was thought to be due to the anti-oxidant activity of the flavonoid (Dajas et al., 2003). In contrast, AF has been shown to activate caspase-3 and disrupt redox balance within cells; therefore, the protective mechanisms of AF are likely different from those of flavonoids (Mirabelli et al., 1985; Jeon et al., 2000; Ellingsen et al., 2001).

77

Ineffectiveness of AF as anti-oxidant has been reported before (Krause et al., 2011). It was confirmed in this study by the lack of the effect of AF on the luminol-dependent chemiluminescence response of HL-60 cells. This assay is dependent on production of ROS by stimulated cells and therefore drugs which are anti-oxidants and can scavenge ROS often inhibit the chemiluminescence signal (Lu et al., 2012; Lee et al., 2013). HL-60 cells were used as surrogates of mononuclear phagocytes that possess enzymatically active NADPH oxidase complexes capable of generating strong respiratory burst response. It is known that mammalian microglia possess this enzymatic activity (Gao et al., 2012) but some cell lines, including human monocytic THP-1 cells, express low levels of NADPH oxidase, which does not allow respiratory burst measurement in these cells (Makino et al., 2012). Low yield of primary human microglia precluded respiratory burst experiments with this cell type. Since a 30 min treatment of HL-60 cells with AF did not inhibit their fMLP-induced luminol-dependent chemiluminescence response, effects of AF on the priming of respiratory burst were studied. Priming of NADPH oxidase is a good target for pharmacological intervention because its inhibition would reduce the pathological increase in ROS production while maintaining the normal physiological NADPH oxidase-mediated immune responses (Muranaka et al., 2005). HL-60 cells were primed with LPS, a known priming agent, and were treated with AF either before the addition of the priming agent or 30 min prior to stimulation with fMLP. The former treatment with AF at the non-toxic 0.1–0.5 μM range inhibited LPS priming of the respiratory burst. Treating HL-60 cells with AF for 30 min did not inhibit the respiratory burst response, which was similar to the lack of the effect of AF on un-primed cells. The extended incubation time needed for the inhibitory effect of AF to appear indicates that AF does not inhibit NADPH oxidase directly; therefore, AF may regulate expression of genes encoding NADPH oxidase components or affect slow-acting cellular pathways (Anderson et al., 1991). Previous research indicates that the majority of AF is taken up by cells within 20 min (Graham et al., 1994); therefore, a 30 min treatment is sufficient for AF to enter cells and interact with NADPH-oxidase subunits if this were the mechanism of AF action (Wong et al., 1990). Agents that inhibit the priming of NADPH-oxidase, but not the functioning of this enzymatic complex, are not common. IL-4 and lipid A analogues have both been shown to inhibit the priming of respiratory burst in neutrophils (Means and Dedman, 1980; Brodie et al., 1998; Chomarat and Banchereau, 1998). IL-4 inhibits NADPH-oxidase assembly by decreasing the expression of its gp91-phox subunit, thus effectively reducing the number of active enzyme complexes (Means and Dedman, 1980; Brodie et al., 1998; Chomarat and Banchereau, 1998). Lipid A analogues are structurally similar to LPS and they are thought to inhibit NADPH-oxidase priming by antagonizing the LPS-binding regions on cell surfaces (Van Dervort et al., 1992). It is possible that AF inhibits priming of the respiratory burst through either of these mechanisms. Future investigations into the exact mechanism by which AF inhibits the LPS-primed respiratory burst should focus on expression of NADPH-oxidase subunits and potential antagonism of LPS receptors. The direct neuroprotective activity of AF has already been reported, and it was shown that up-regulation of HOX-1 could contribute to this protective activity in neuronal SH-SY5Y cells (Madeira et al., 2013). Previous studies by Kim et al. (2010) have shown that AF upregulates HOX-1 in THP-1 cells, therefore upregulation of HOX-1 could be responsible for the anti-inflammatory effects of AF in different types of mononuclear phagocytes, including microglia. HOX-1 has been previously suggested as a protective molecule in neuroinflammatory conditions. Yamamoto et al. (2010) demonstrated that induction of HOX-1 protected dopaminergic neurons in an in vitro model of Parkinson's disease. By up-regulating protective enzymes like HOX-1, AF could directly increase the ability of neurons to withstand toxic insults. Our study identifies potential anti-neurotoxic and neuroprotective effects of AF in microglia-like cells. Previous studies of the in vivo

78

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79

distribution of AF after its oral administration in rodents (1–2 mg/kg per day for 5–7 days) have shown that the CNS concentration of AF could reach 0.2–5 μM range (Walz et al., 1983; Madeira et al., 2013). This indicates that AF can cross the blood–brain barrier and may reach anti-neurotoxic and neuroprotective concentrations in the CNS. Furthermore, the blood–brain barrier becomes compromised in neurodegenerative diseases, such as Alzheimer's and Parkinson's diseases (Kortekaas et al., 2005; Persidsky et al., 2006), which could lead to even higher brain concentrations of AF after its oral administration. Therefore, further testing of AF in animal models of neuroinflammation is warranted. Extensive clinical use of AF has shown that this drug has a relatively safe pharmacological profile in humans including elderly patients (Glennas et al., 1997; Kean et al., 1997; Kim et al., 2010). This, in combination with its anti-inflammatory and cytoprotective activity demonstrated by this study may make AF a good treatment option for inflammatory conditions other than rheumatoid arthritis. Antiinflammatory and neuroprotective treatment strategies have the potential to slow the progression of neurological disorders including Alzheimer's and Parkinson's diseases and should be explored in clinical studies since currently there is no effective treatment for these diseases (Klegeris et al., 2007; Heneka et al., 2010).

Acknowledgments This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada and the Jack Brown and Family Alzheimer's Disease Research Foundation. We would like to thank Mr. C.J. Renschler for his technical assistance.

References Abramson, S.L., Gallin, J.I., 1990. IL-4 inhibits superoxide production by human mononuclear phagocytes. J. Immunol. 144, 625–630. Anderson, R., Van Rensburg, C.E., Joone, G.K., Lessing, A., 1991. Auranofin inactivates phosphofructokinase in human neutrophils, leading to depletion of intracellular ATP and inhibition of superoxide generation and locomotion. Mol. Pharmacol. 40, 427–434. Block, M.L., Zecca, L., Hong, J.S., 2007. Microglia-mediated neurotoxicity: uncovering the molecular mechanisms. Nat. Rev. Immunol. 8, 57–69. Brodie, C., Goldreich, N., Haiman, T., Kazimirsky, G., 1998. Functional IL-4 receptors on mouse astrocytes: IL-4 inhibits astrocyte activation and induces NGF secretion. J. Neuroimmunol. 81, 20–30. Brown, G., 2010. Nitric oxide and neuronal death. Nitric Oxide 23, 153–165. Bruze, M., Björkner, B., Möller, H., 1995. Skin testing with gold sodium thiomalate and gold sodium thiosulfate. Contact Dermatitis 32, 5–8. Cameron, B., Landreth, G.E., 2010. Inflammation, microglia, and Alzheimer's disease. Neurobiol. Dis. 37, 503–509. Champion, G.D., Graham, G.G., Ziegler, J.B., 1990. The gold complexes. Baillieres Clin. Rheumatol. 4, 491–534. Chomarat, P., Banchereau, J., 1998. Interleukin-4 and interleukin-13: their similarities and discrepancies. Int. Rev. Immunol. 17, 1–52. Cornelius, C., Crupi, R., Calabrese, V., Graziano, A., Milone, P., Pennisi, G., Radak, Z., Calabrese, E.J., Cuzzocrea, S., 2013. Traumatic brain injury: oxidative stress and neuroprotection. Antioxid. Redox Signal. 19, 836–853. Dajas, F., Rivera, F., Blasina, F., Arredondo, F., Echeverry, C., Lafon, L., Morquio, A., Heinzen, H., 2003. Cell culture protection and in vivo neuroprotective capacity of flavonoids. Neurotox. Res. 5, 425–432. Decker, T., Lohmann-Matthes, M.L., 1988. A quick and simple method for the quantitation of lactate dehydrogenase release in measurements of cellular cytotoxicity and tumor necrosis factor (TNF) activity. J. Immunol. Methods 115, 61–69. Ding, A.H., Nathan, C.F., Stuehr, D.J., 1988. Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages. Comparison of activating cytokines and evidence for independent production. J. Immunol. 141, 2407–2412. Ellingsen, T., Buus, A., Stengaard-Pedersen, K., 2001. Plasma monocyte chemoattractant protein 1 is a marker for joint inflammation in rheumatoid arthritis. J. Rheumatol. 28, 41–46. Frohman, E.M., Racke, M.K., Raine, C.S., 2006. Multiple sclerosis — the plaque and its pathogenesis. N. Engl. J. Med. 354, 942–955. Gao, H.M., Zhou, H., Hong, J.S., 2012. NADPH oxidases: novel therapeutic targets for neurodegenerative diseases. Trends Pharmacol. Sci. 33, 295–303. Gibbons, H., Dragunow, M., 2006. Microglia induce neural cell death via a proximitydependent mechanism involving nitric oxide. Brain Res. 1084, 1–15.

Glennas, A., Kvien, T.K., Andrup, O., Clarke-Jenssen, O., Karstensen, B., Brodin, U., 1997. Auranofin is safe and superior to placebo in elderly-onset rheumatoid arthritis. Br. J. Rheumatol. 36, 870–877. Graham, G.G., Champion, G.D., Ziegler, J.B., 1994. The cellular metabolism and effects of gold complexes. Metal-Based Drugs 1, 395–404. Hansen, M.B., Nielsen, S.E., Berg, K., 1989. Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J. Immunol. Methods 119, 203–210. Hashioka, S., Klegeris, A., McGeer, P.L., 2009. Proton pump inhibitors exert antiinflammatory effects and decrease human microglial and monocytic THP-1 cell neurotoxicity. Exp. Neurol. 217, 177–183. Heneka, M.T., O'Banion, M.K., Terwel, D., Kummer, M.P., 2010. Neuroinflammatory processes in Alzheimer's disease. J. Neural Transm. 117, 919–947. Jeon, K.I., Jeong, J.Y., Jue, D.M., 2000. Thiol-reactive metal compounds inhibit NF-kappa B activation by blocking I kappa B kinase. J. Immunol. 164, 5981–5989. Kean, W.F., 1990. Intramuscular versus oral gold therapy. Baillieres Clin. Rheumatol. 4, 219–246. Kean, W.F., Hart, L., Buchanan, W.W., 1997. Auranofin. Br. J. Rheumatol. 36, 560–572. Kim, N.H., Lee, M.Y., Park, S.J., Choi, J.S., Oh, M.K., Kim, I.S., 2007. Auranofin blocks interleukin-6 signalling by inhibiting phosphorylation of JAK1 and STAT3. Immunology 122, 607–614. Kim, N.H., Oh, M.K., Park, H.J., Kim, I.S., 2010. Auranofin, a gold (I)-containing antirheumatic compound, activates Keap1/Nrf2 signaling via Rac1/iNOS signal and mitogen-activated protein kinase activation. J. Pharmacol. Sci. 113, 246–254. Klegeris, A., Bissonnette, C.J., McGeer, P.L., 2003. Reduction of human monocytic cell neurotoxicity and cytokine secretion by ligands of the cannabinoid-type CB2 receptor. Br. J. Pharmacol. 139, 775–786. Klegeris, A., Bissonnette, C.J., McGeer, P.L., 2005. Modulation of human microglia and THP-1 cell toxicity by cytokines endogenous to the nervous system. Neurobiol. Aging 26, 673–682. Klegeris, A., McGeer, E.G., McGeer, P.L., 2007. Therapeutic approaches to inflammation in neurodegenerative disease. Curr. Opin. Neurol. 20, 351–357. Kortekaas, R., Leenders, K.L., van Oostrom, J.C., Vaalburg, W., Bart, J., Willemsen, A.T., Hendrikse, N.H., 2005. Blood–brain barrier dysfunction in Parkinsonian midbrain in vivo. Ann. Neurol. 57, 176–179. Krause, D., Suh, H.S., Tarassishin, L., Cui, Q.L., Durafourt, B.A., Choi, N., Bauman, A., CosenzaNashat, M., Antel, J.P., Zhao, M.L., Lee, S.C., 2011. The tryptophan metabolite 3-hydroxyanthranilic acid plays anti-inflammatory and neuroprotective roles during inflammation: role of hemeoxygenase-1. Am. J. Pathol. 179, 1360–1372. Lee, Y.J., Han, S.B., Nam, S.Y., Oh, K.W., Hong, J.T., 2010. Inflammation and Alzheimer's disease. Arch. Pharm. Res. 33, 1539–1556. Lee, C.W., Ko, H.H., Lin, C.C., Chai, C.Y., Chen, W.T., Yen, F.L., 2013. Artocarpin attenuates ultraviolet B-induced skin damage in hairless mice by antioxidant and antiinflammatory effect. Food Chem. Toxicol. 60, 123–129. Li, C., Zhao, R., Gao, K., Wei, Z., Yin, M.Y., Lau, L.T., Chui, D., Hoi Yu, A.C., 2011. Astrocytes: implications for neuroinflammatory pathogenesis of Alzheimer's disease. Curr. Alzheimer Res. 8, 67–80. Lu, D., Shao, H.T., Ge, W.P., Liu, N., Zhang, X., Ma, C.M., Qin, C., Zhang, L.F., 2012. Ginsenoside-RB1 and tetramethylpyrazine phosphate act synergistically to prevent dilated cardiomyopathy in cTnTR141W transgenic mice. J. Cardiovasc. Pharmacol. 59, 426–433. Lull, M.E., Block, M.L., 2010. Microglial activation and chronic neurodegeneration. Neurotherapeutics 7, 354–365. Madeira, J.M., Gibson, D.L., Kean, W.F., Klegeris, A., 2012a. The biological activity of auranofin: implications for novel treatment of diseases. Inflammopharmacology 20, 297–306. Madeira, J.M., Little, J.P., Klegeris, A., 2012b. Microglia secretome: from neurotoxins to neurotrophins. In: Kaur, C., Ling, E.A. (Eds.), Microglia: Biology, Functions and Roles in Disease. Nova Publishers, New York, pp. 73–92. Madeira, J.M., Renschler, C.J., Mueller, B., Hashioka, S., Gibson, D.L., Klegeris, A., 2013. Novel protective properties of auranofin: inhibition of human astrocyte cytotoxic secretions and direct neuroprotection. Life Sci. 92, 1072–1080. Makino, J., Kamiya, T., Hara, H., Adachi, T., 2012. TPA induces the expression of EC-SOD in human monocytic THP-1 cells: involvement of PKC, MEK/ERK and NOX-derived ROS. Free Radic. Res. 46, 637–644. Means, A.R., Dedman, J.R., 1980. Calmodulin—an intracellular calcium receptor. Nature 285, 73–77. Mirabelli, C.K., Johnson, R.K., Sung, C.M., Faucette, L., Muirhead, K., Crooke, S.T., 1985. Evaluation of the in vivo antitumor activity and in vitro cytotoxic properties of auranofin, a coordinated gold compound, in murine tumor models. Cancer Res. 45, 32–39. Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63. Muranaka, S., Fujita, H., Fujiwara, T., Ogino, T., Sato, E.F., Akiyama, J., Imada, I., Inoue, M., Utsumi, K., 2005. Mechanism and characteristics of stimuli-dependent ROS generation in undifferentiated HL-60 cells. Antioxid. Redox Signal. 7, 1367–1376. Nakaya, A., Sagawa, M., Muto, A., Uchida, H., Ikeda, Y., Kizaki, M., 2011. The gold compound auranofin induces apoptosis of human multiple myeloma cells through both downregulation of STAT3 and inhibition of NF-kappaB activity. Leuk. Res. 35, 243–249. Persidsky, Y., Ramirez, S.H., Haorah, J., Kanmogne, G.D., 2006. Blood–brain barrier: structural components and function under physiologic and pathologic conditions. J. Neuroimmune Pharmacol. 1, 223–236. Pocernich, C.B., Butterfield, D.A., 2012. Elevation of glutathione as a therapeutic strategy in Alzheimer disease. Biochim. Biophys. Acta 1822, 625–630. Polanski, W., Reichmann, H., Gille, G., 2011. Stimulation, protection and regeneration of dopaminergic neurons by 9-methyl-β-carboline: a new anti-Parkinson drug? Expert. Rev. Neurother. 11, 845–860.

J.M. Madeira et al. / Journal of Neuroimmunology 276 (2014) 71–79 Ransohoff, R.M., Brown, M.A., 2012. Innate immunity in the central nervous system. J. Clin. Invest. 122, 1164–1171. Stern, I., Wataha, J.C., Lewis, J.B., Messer, R.L., Lockwood, P.E., Tseng, W.Y., 2005. Antirheumatic gold compounds as sublethal modulators of monocytic LPS-induced cytokine secretion. Toxicol. in Vitro 19, 365–371. Van Dervort, A.L., Doerfler, M.E., Stuetz, P., Danner, R.L., 1992. Antagonism of lipopolysaccharide-induced priming of human neutrophils by lipid A analogs. J. Immunol. 149, 359–366. Walz, D.T., DiMartino, M.J., Griswold, D.E., Intoccia, A.P., Flanagan, T.L., 1983. Biologic actions and pharmacokinetic studies of auranofin. Am. J. Med. 75, 90–108. Wong, K., Parente, J., Prasad, K.V., Ng, D., 1990. Auranofin modulated cytoplasmic free calcium in neutrophils by mobilizing intracellular calcium and inhibiting protein kinase. J. Biol. Chem. 265, 21454–21461. Yamada, R., Sano, H., Hla, T., Hashiramoto, A., Fukui, W., Miyazaki, S., Kohno, M., Tsubouchi, Y., Kusaka, Y., Kondo, M., 1999. Auranofin inhibits interleukin-1betainduced transcript of cyclooxygenase-2 on cultured human synoviocytes. Eur. J. Pharmacol. 385, 71–79.

79

Yamamoto, N., Izumi, Y., Matsuo, T., Wakita, S., Kume, T., Takada-Takatori, Y., Sawada, H., Akaike, A., 2010. Elevation of heme oxygenase-1 by proteasome inhibition affords dopaminergic neuroprotection. J. Neurosci. Res. 88, 1934–1942. Yamashita, M., Niki, H., Yamada, M., Watanabe-Kobayashi, M., Mue, S., Ohuchi, K., 1997. Dual effects of auranofin on prostaglandin E2 production by rat peritoneal macrophages. Eur. J. Pharmacol. 325, 221–227. Zhang, W., Wang, T., Pei, Z., Miller, D.S., Wu, X., Block, M.L., Wilson, B., Zhang, W., Zhou, Y., Hong, J.S., Zhang, J., 2005. Aggregated alpha-synuclein activates microglia: a process leading to disease progression in Parkinson's disease. FASEB J. 19, 533–542. Zilka, N., Kazmerova, Z., Jadhav, S., Neradil, P., Madari, A., Obetkova, D., Bugos, O., Novak, M., 2012. Who fans the flames of Alzheimer's disease brains? Misfolded tau on the crossroad of neurodegenerative and inflammatory pathways. J. Neuroinflammation 9, 47.